For silyl ketene acetals, Petrov et al. first reported their synthesis (see J. Gen. Chem. (USSR), 1959, vol. 29, pp. 2896-2899). The silyl ketene acetals are compounds of great commercial interest. One of their applications is the use as polymerization initiators in the polymerization of acrylate monomers, known as “group transfer polymerization,” developed by Webster et al. (see U.S. Pat. No. 4,417,034 and U.S. Pat. No. 4,508,880). Another application is the use as nucleophilic agents in the synthesis of carboxylic acid derivatives (see JP-A 2001-247514).
Disilyl ketene acetals are regarded as one class of silyl ketene acetals from the standpoint of chemical structure. Therefore, most of their preparation processes are in accord with processes of silyl ketene acetal preparation.
Heretofore, for the preparation of silyl ketene acetals, four predominant processes are known in the art. They are (1) the reaction of a carboxylic ester having a hydrogen atom at α-position with a base and a silylating agent; (2) the activation of a carboxylic ester having a halogen atom substituted at α-position with a metal such as sodium or zinc, followed by reaction with a silylating agent such as chlorotrimethylsilane; (3) the reaction of a malonic ester with a silylating agent such as chlorotrimethylsilane in the presence of metallic sodium; and (4) the reaction of an α,β-unsaturated carboxylic ester with a hydrosilane or hydrosiloxane in the presence of a transition metal catalyst.
In process (1), typical combinations of base/silylating agent include lithium diisopropylamide/chlorotrimethylsilane (see JP-A 9-221444, for example) and triethylamine/trimethylsilyl trifluoromethanesulfonate (see U.S. Pat. No. 4,482,729, for example). In either case, reaction proceeds at room temperature or lower temperatures, but requires to use at least one equivalent of the base, forming a large amount of salt. This is detrimental particularly when the process is applied to a large scale of production. The latter combination sometimes results in low yields with certain substrates because a compound having alpha-carbon silylated is produced in addition to the desired silyl ketene acetal.
In process (2) as exemplified by JP-A 2-111780 and process (3) as exemplified by JP-A 64-85982, at least one equivalent of a metal such as sodium is used, forming a large amount of metal salt. The metal salt must be removed before the desired silyl ketene acetal can be isolated. Also, since the metal is often used in excess, the metal salt formed contains metal in the activated state, requiring careful handling. Thus these processes are difficult to implement on a large scale.
Unlike processes (1) to (3), process (4) utilizes addition reaction, offering the advantage that no waste products like the above-mentioned salt are formed. It is known that this process uses transition metal compounds as the catalyst. In the article of Petrov et al., for example, a platinum compound is used as the catalyst. Among others, rhodium catalysts are effective. For instance, chlorotris(triphenylphosphine)rhodium is used in Chem. Pharm. Bull., 1974, vol. 22, pp. 2767-2769 and JP-A 63-290887 and rhodium trichloride trihydrate used in JP-A 62-87594. In U.S. Pat. No. 5,208,358 a successful use of chlorobis(di-tert-butylsulfide)rhodium as a catalyst has been disclosed in producing silyl ketene acetals with lower catalyst loadings. However, as described in JP-A 62-87594, when hydrosilylation reaction is catalyzed by transition metal catalysts, there are formed not only the desired silyl ketene acetal, but also by-products such as carbonyl adducts or β-adducts which have a boiling point close to the desired product and are difficult to separate by distillation. It is then difficult to obtain silyl ketene acetals with high purity. In JP-A 62-87594, a silyl ketene acetal is obtained in a highly pure form (≧95%) by using rhodium trichloride trihydrate as the catalyst and an excess amount of hydrosilane and converting the carbonyl adduct to a high-boiling compound. However, by this process, it was difficult to obtain silyl ketene acetals in high yields because of the decreased yields based on the hydrosilane used, and of the increased distillation residue leading to lower isolated yields.
Thus, the process of making silyl ketene acetals by hydrosilylation of α,β-unsaturated carboxylic esters is advantageous in that it does not essentially generate by-products such as salts. Nonetheless, with conventional transition metal catalysts, it suffers from the low selectivity owing to the formation of multiple products derived from a variety of hydrosilylation modes. It would be desirable to have a process of preparing a silyl ketene acetal at a good selectivity, high purity and high yield.
For the preparation of disilyl ketene acetals, any of the foregoing processes is applicable. The preparation of disilyl ketene acetals generally starts with silyl carboxylic esters, most of which are not commercially available. This necessitates the extra step of preparing silyl carboxylates beforehand through silylation of carboxylic acids, adding to the cost of manufacture.
A process other than stated above has been proposed for disilyl ketene acetal manufacturing: a reaction between α,β-unsaturated carboxylic esters and hydrosilanes in the presence of a rhodium catalyst (see JP-A 64-71886). With this process, disilyl ketene acetals can be obtained in one step using allyl esters of α,β-unsaturated carboxylic acids such as commercially available allyl methacrylate. Although the reaction mixture resulting from this process is allegedly free of a typical by-product, carbonyl adduct, the yield and purity of the desired product are below satisfactory levels because other by-products are formed in noticeable amounts.
The existing processes for the preparation of disilyl ketene acetals suffer from drawbacks such as large amounts of by-product generation and laborious purification of the desired compounds. It would be desirable to have a process of preparing disilyl ketene acetals in high purity and high yields without forming substantial by-products.